Identification of novel mechanistic markers with prognostic value in clinical course of chronic kidney disease (CKD) will provide a better understanding of disease process and improve care of the patients by identification of high risk groups at earlier stages. Importantly, identification of mechanistic pathways will be amenable to novel preventive or therapeutic interventions. This study is aimed at comparing the predictive power of ANCUAL (Amino acids, Nucleotides, intermediates of Citric acid and Urea cycles, Acylcarnitines, and Lipids) metabolic panel to predict progression of CKD by comparing their area under Receiver Operating Characteristics (ROC) curves with those of the traditional markers (serum creatinine, urinary protein, eGFR) at early stage. We also aim to explore the link between the proposed metabolic markers with the renal tissue expression level of genes involved in regulation of these markers. Methods: This will be a nested case-cohort study. Study populations consist of participants of Chronic Renal Insufficiency Cohort (CRIC) for measurement of metabolites, and Clinical Phenotyping Resource and Biobank Core (CPROBE) for transcriptomics- metabolomics integrative analysis. Inclusion criteria are entry age of 21 to 74 years with eGFR of 30-70 for CRIC participants, and entry eGFR of 30 mL/min or higher for CPROBE participants. Exclusion criteria are the same as CRIC and CPROBE core studies. Sampling: From CRIC participants; case group is defined as 100 patients who have progressed to ESRD including 50 patients with diabetes and 50 patients without diabetes. Control group will be 100 patients matched by age, sex, race, diabetes, and baseline eGFR (N=200). This sampling will be replicated to get another 200 patients from CRIC in the validation phase (total=400). From CPROBE participants; 42 participants with available gene expression data obtained from kidney biopsy will be selected. Data collection: Starting with CRIC, fasting serum samples from baseline will be collected. First, samples will be subjected to targeted metabolomics analysis by mass spectrometry (MS) for quantification of ANCUAL panel, in the first 200 CRIC participants. This will be followed by an untargeted metabolomic analysis for expansion of the initial panel, followed by an orthogonal quantification of the newly identified prognostic metabolites to validate the discovery of the new compounds from untargeted platform. The expanded ANCUAL panel will then undergo replication in a targeted platform in another 200 independent samples from CIRC to validate the final panel. At last step the final panel will be quantified in the baseline serum samples of CPROBE participants in an attempt to integrate with their transcriptomics data. Predictors are the expanded ANCUAL panel derived from MS. Outcome is incident ESRD. Covariates include demographics, laboratory values, comorbidities, physical examination findings and anthropometric measures from the time of sample collection. Analysis: Multiple logistic regression analysis will be applied to identify the independent components of a multi panel marker among the differentially expressed baseline panel. C-statistics will be applied to compare the area under Receiver Operating Characteristics (ROC) curve of the multi-panel marker with that if the traditional biomarkers for discrimination of progressors from non-progressors. Integrated discrimination improvement, net reclassification index, and C-statistics will be applied to assess the classification improvement of the expanded panel compared to baseline panel. Risk of ESRD by quartiles of the proposed panel will be estimated using logistic regression adjusting for relevant prognostic covariates. Gene and metabolic set enrichment analysis will be applied to test enrichment of differentially regulated metabolic pathways. Bioinformatics tools will be used for pathway recognition.